传统的线激光扫描技术以机器视觉为基础,算法复杂、计算量大,且匹配效率不高;而点激光扫描技术是以激光三角法为基础,算法简洁,但测量慢、点云稀疏。为解决上述问题,提出以激光三角法为基础的线激光扫描技术,根据相机针孔模型以及相机和线激光器的相对位置关系建立物像关系方程,求解场景三维坐标,简化了重建算法。采用以最小二乘法为基础的标定算法,实现了相机与线激光相对位置参数和系统旋转中心偏移参数的标定。同时分析了系统的理论精度,并结合场景重建实验和精度评估实验,对系统的可靠性和精度进行了验证。实验结果表明该系统效率高、精度好;在1000~1700 mm测量范围内该系统的绝对误差小于2.6 mm,精度高于0.25%,满足一般场景重建的要求。

The traditional line laser scanning technology is based on machine vision, the corresponding algorithm is complex, the amount of calculation is large, and the matching efficiency is not high. In contrast, the point laser scanning technology is based on laser triangulation, the corresponding algorithm is concise, but the measurement speed is low and the reconstructed point cloud is sparse. To solve these problems, the laser-triangulation-based line laser scanning technology is proposed. The camera pinhole model and the relative position between camera and line laser are used for the establishment of an object-image relationship equation. The 3D coordinates of the scene are solved and the whole reconstruction algorithm is simplified. The calibration algorithm based on the least square method is adopted for the calibration of parameters such as the relative position between camera and line laser and the rotation center deviation of the system. Simultaneously, the theoretical precision of the system is analyzed, and the reliability and accuracy are tested by the experiments of scene reconstruction and accuracy evaluation. The experimental results show that the system possesses a high efficiency and a good precision. The error of the system is less than 2.6 mm and the precision is higher than 0.25% in the measurement range of 1000-1700 mm, which satisfies the requirements for a general scene reconstruction.